4.6 Workers’ Compensation - WordPress.com · SAS has been partnering with customers to solve...

Tab 4—Subject Matter Categories/Areas (Appendix A)

A SAS Solution for Enterprise Data Analytics | 179

4.6 Workers’ Compensation A. A description of relevant experience specific to this subject matter category;

For the past 10 years, SAS has been developing and deploying solutions to help our

customers detect, manage and prevent fraud. Our proposal outlines the reasons that

substantiate SAS and the SAS Fraud Framework’s ability to meet the State’s

requirements now and in the future.

SAS Solutions OnDemand’s key responsibility is to deliver the SAS Fraud

Framework for our fraud customers, whether hosted by SAS or the customer. Nearly

150 customers trust SAS Solutions OnDemand with their solutions. All of SAS’

global fraud resources are available to support SAS Solutions OnDemand in

ensuring a successful project for the State. Successful customer projects include the

following:

Los Angeles County, CA—expects return on investment of between $7

million and $30 million annually and has achieved an 85 percent accuracy in

identification of suspected fraud rings.

Royal Bank of Canada—uses SAS to build a fraud rules application that

applies powerful analytics to target suspect applications.

BBVA Bancomer—uses SAS Fraud Detection and Prevention for Financial

Services to detect suspicious credit/debit card activity with greater ease and

precision.

Commonwealth Bank of Australia—uses SAS to improve its fraud detection

efficiency and has detected twice the level of check fraud, increased Internet

banking fraud alerts by 60 percent and had a 60 percent improvement in

Internet banking fraud-alert volumes using SAS.

Health Care Services Corporation—saves millions of dollars using SAS to

identify fraudulent healthcare claims before they are ever paid.

Highmark—uses SAS to stop millions of dollars in fraudulent health

insurance claims from being paid.

Louisiana Workforce Commission—expects 8-to-1 ROI and has already

received $450,000 from one of the first companies identified.

Medical Benefits Fraud (Australia's largest privately owned insurance

company)—relies on SAS to create an anti-fraud culture and realized a 200

percent return on investment.

PBS (the leading provider of electronic payment systems in Denmark)—uses

SAS to improve processing time and employ better analytic tools to fight

credit card fraud.

Washington State Department of Labor and Industries expects an 8-to-1

return on investment as has seen an 80% increase in efficiency using the

SAS Fraud Framework.

WorkCover, the workers’ compensation program in New South Wales

Australia, applied the Employer Premium Payments model, resulting in a

10 years of fraud

specific solutions

development – plus 24

years of foundational

technology development

SAS Solutions

OnDemand customers

span the global

spectrum

“SAS is committed to

ensuring that we

continue to have a

leading-edge anti-fraud

solution. We are very

pleased with the results.”

- Derek Wylde

Head of Group Fraud

Risk, Global Security

and Fraud Risk

HSBC Bank



savings of AUD$3 million (US$1.98 million). Since applying the model, there

has been a doubling in collected premiums for two consecutive years and

net revenue of audits has increased fivefold.

State of North Carolina Department of Justice—Although not fraud, uses

SAS to consolidate information across all criminal justice entities to create an

offender centric view of all offender information. Data sources span multiple

branches and agencies of state government, local courts, jails and arrest

data, and select federal information. The system provides robust search and

watch capabilities so that any criminal justice professional can get accurate

and timely offender information at the touch of a button.

B. A description of the data analytics services you are proposing to offer in

this subject matter category (be specific, noting that agencies may seek to

contract for a subset of the offered services);

SAS is dedicated to your long-term success. The intelligence that is housed in your

data is unique, defining the diverse facets of who you are, what you have

accomplished, and how you can succeed. At SAS, we are committed to helping you

transform this raw data into vital enterprise intelligence by delivering the technical

solutions and services you need. We blend award-winning software with flexible

business and technical consulting services to turn your business strategies into a real

advantage.

SAS has been partnering with customers to solve their business problems for more

than 35 years. SAS consultants take the time to listen and learn about your business

challenges and enterprise goals to establish a foundation for strategic advancement.

This enables SAS to deliver the right SAS technology and tailored services to solve

your unique business requirements.

SAS consultants have a wealth of experience gained working with many hundreds of

customers across all industry sectors including finance, telecommunications, retail,

and government. Alongside this industry domain expertise, we give customers

practical insights into how their organization can use business analytics to become

more focused, more effective and ultimately more successful. Such advice can be

offered independent of any SAS software.

SAS Professional Services provides expert advice and help at all stages in a

project’s life cycle. Specific services range from solution discovery workshops,

through to cleaning and analyzing data, evaluating your architecture, and

implementing the right business solution to address customers’ needs. S AS

provides all the help customers need to drive project success.

SAS’ methodologies allow for seamless integration of multi-disciplinary teams to

achieve a more efficient and effective use of resources with better alignment of the

SAS combines expert

technical consulting

services with award-

winning software to

ensure long-term

success

Domain expertise

across all industry

sectors

Solving your unique

business requirements

PMM

Range of services



specialist strands. Along with the technical implementation or solution methodologies

is the Project Management Methodology (PMM). PMM is used for all

implementations to provide a consistent and unified project approach regardless of

the number of solutions or focus of the project.

Our expert SAS Analytical Consultants are available to share their vast knowledge

and real world experience to address:

The analytical approach to a particular business problem.

Additional insight and foresight that could be gleaned from your data.

Building a high performance enterprise that is driven by analytics.

Our goal is your ultimate satisfaction with SAS software and to become your trusted

technology and business partner. To support your analytical needs, SAS offers a

variety of customized or defined consulting services options.

The SAS Analytic Workshop begins with a discovery session between your business

professionals, analytics experts and seasoned SAS analytic consultants. We discuss

your specific business needs and how SAS data mining and forecasting tools and

solutions address those needs. The workshop will help you identify and explore

alternatives to expand the value of your current SAS analytical software.

The SAS Installation & Configuration Service provides a proven process for quickly

and properly installing your SAS analytical software and configuring it to run

effectively within your environment. We verify the software was installed correctly,

functions properly and integrates with other SAS software products and platforms.

The SAS Analytic QuickStart enables your analytical experts to work hand-in-hand

with SAS Analytical Consultants as SAS Enterprise Miner™ or Text Miner is

deployed within your organization. SAS Analytical Consultants will design a custom

knowledge transfer session to help your analytical specialist quickly learn and utilize

the SAS products to meet your specific project’s needs and review best practices for

the construction of flow diagrams in SAS Enterprise Miner™ to build predictive

models.

The SAS Data Mining Service allows you to leverage powerful SAS analytic software

and expert consultants to identify previously unknown patterns and trends in your

data that could impact your organization. During the service, we teach you data

modeling techniques, develop a data mining model using your own data and train

your staff in analytical techniques, the use of SAS Enterprise Miner™ software and

industry best practices, allowing you to become self-sufficient in these areas.

Analytic Workshop

Installation and

Configuration

Analytic QuickStart

Data Mining Service



The SAS Text Mining Service leads you through a systematic process to define your

text mining business objectives. Using your data, we introduce you to our text mining

product by demonstrating text pre-processing, clustering and predictive modeling.

Similarities and patterns are identified in profile clusters, providing you keys to unlock

valuable data.

The SAS Analytic Modeling Service enables SAS Analytical Consultants to review

available data sources to identify possible variables of interest for extraction.

The data sources will be inspected for general data cleanliness and information

content that can be used in model development. SAS Analytical Consultants will

identify driver variables and develop models using appropriate statistical methods to

address the customer’s business need and score the data for use in production.

The SAS Forecasting Service offers you the ability to leverage SAS’ powerful

forecasting capabilities through data management, analytics and reporting with

expert SAS assistance. SAS Analytical Consultants will help you understand trends,

seasonality, events and behavioral characteristics of historical data. SAS Analytical

Consultants will educate your staff about both analytic techniques and forecasting,

enabling you to embark on the path toward self-sufficiency in these areas.

The SAS Customer Care Program provides you the ability to purchase a block of

consulting hours and use them over a pre-determined timeframe. If you have

analytical experts in-house, you can use these hours to access analytical experts at

SAS with whom you can discuss model development, statistical methods, or any

other analytical needs. If you don’t have analytical experts in-house, you can use

these hours to leverage SAS Analytical Consultants to assist you in building models,

maximizing use of your SAS analytical software and knowledge transfer of analytical

concepts including industry best practices.

SAS has been listening to our customers for 35 years, and that is why we understand

your business, competition and IT environment. No one knows SAS technology

better than SAS consultants. Our hands-on experience—drawn from more than

1,500 expert consultants worldwide—provides your organization with the fastest

proven path to successful implementations of SAS Solutions. Our flexible

engagement model is structured to meet your specific business and industry needs.

While we work directly with your project team, we also serve as your liaison to SAS

Research and Development and Technical Support. Your success is our shared

success.

Service Levels (Levels of sophistication or maturity)

1. Reporting/Information Extraction

• Operational reporting

• Parameter driven reports

Text Mining Service

Analytic Modeling

Service

Forecasting Service

Customer Care

Program

Listening to our

customers



• Pre-defined Queries

• Pre-defined Data Cube(s) and/or Mart(s)

• Pivot Tables

Reporting/Information Extraction are key services provided by the SAS team using

the SAS Fraud Framework. SAS consultants will develop the fraud and abuse

performance report portfolio. These reports include operational reporting, parameter

driven reports, pre-defined queries, pre-defined data cube(s) and mart(s) and pivot

tables. In addition, ad hoc reports are available.

2. Data Analysis

• Direct Access to Data

• Economic and Financial Analysis

• Data Cube(s) and/or Mart(s) creation

• Ad hoc Queries

• On Line analytical processing (OLAP)

• Use of Business Intelligence (BI) tools

Data analysis is another key service provided by the SAS team using the SAS Fraud

Framework. Some of the services offered by SAS consultants include: Establish an ETL process for direct access and/or mapping to source data

feeds for existing and new data sources

Establish a production batch schedule for existing and new data sources

Integrate existing business rules from various sources

Develop fraud and abuse management dashboard

Implement sophisticated name matching and record linking techniques

(including fuzzy matching) to match entities across and within data sources

In implementing the SAS Fraud Framework solution to address workers’

compensation fraud, specific rules and analytical models (anomaly detection and

predictive models) will be used to maximize the prevention of fraud. The analytics

and data used in each solution will be shared accordingly to leverage the fraud

detection capability across the platform. The Social Network Analysis capability will

also be utilized across all of these initiatives to facilitate the sharing of fraud

information across programs.

By combining the disparate data sources, running fraud detection analytics, and

applying social network analysis, a more complete picture of employees, employers,

and service providers will emerge allowing for:

Detection of previously unknown claims fraud

Detection of previously unknown employer and service provider fraud and

abuse

Increased referrals for legal and civil penalties

Increased referrals for prosecution

Maximize prevention of

fraud with the SAS

Fraud Framework’s

integrated advanced

analytics

Creating a more

complete picture of

employees, employers

and service providers

In its 2010 Forrester

Wave report, SAS was

cited as a leader in

Enterprise Business

Intelligence Platforms

SAS/DataFlux is

strongly positioned in

the visionary quadrant

of the Gartner Magic

Quadrant for Data

Integration Tools, 2010



Sentinel deterrent effect of imposing penalties and prosecution on would-be

fraudsters

Increased amounts of recovered monies

Lower insurance premiums for Minnesota businesses

A level playing field for honest businesses

A more attractive business climate to recruit and retain businesses

Workers’ Compensation Premium Evasion

Employers misclassify workers to avoid withholding income taxes, paying

employment taxes, and workers’ compensation premiums. Premium evasion fraud

comes in two forms. First is the intentional misreporting and underreporting of

employees and employee hours. Second, there are employers who operate in an

underground economy and don’t report any or a sizable portion of their workers.

Based upon our experience, to maximize detection, integration of several key data

sources is recommended. These data sources will allow the State to obtain a single

view of the entities and understand behavior holistically to better profile for fraud.

This is NOT a complete list, but is representative of the types of data that have been

useful in other engagements:

Employer database—static data on the businesses receiving workers’

compensation benefits

• Contact information (e.g., address, phone, email, industry data)

• Owner information (e.g., static data of owners/spouses, ownership

dates and %)

• Place of business—all details of the respective places of business

Policy data—specific details on coverage purchased/subscribed to by

employers and any changes to the coverage

Employer premium reports—employer filings of hours worked and worker

risk class (used to derive premiums owed)

Workers’ compensation claims—claim data including injured party, injury

details (e.g., date/time, description, work type, witness), employment history,

etc.

Accounts receivables—provides data on accounts receivables for premiums

and other agency receivables

Premium payments—data on accounts being used to pay premiums

Safety inspection data—can be used to compare the workers on a site to

what is being reported

Wage claim data—employees that have filed a claim for unpaid wages

Fraud/audit case details—outcomes and information from past fraud

investigations and audits

Referral information—details of all fraud and audit referrals from internal and

external sources

Data sources make up

the foundation of the

solution



Department of Licensing/Contractor Registration—registration data for

employer licenses and contractors (can also leverage Secretary of State)

Department of Revenue—total revenue reporting for employers/businesses

Building permits—details of the permits and type of work being completed by

businesses

Office of Motor Vehicles—details of individuals registered with the state

Federal and State Directory of New Hires—static and employment

information on recently hired workers

IRS data—details of business and individual returns

Known bad lists—hot lists of employers that have been flagged for past fraud

3rd Party commercial databases—business/employer, fraud hot lists, and

fraud scheme details

SAS will leverage data sources in the State which include any suspicious activities

such as known bad lists or public referrals. Upon integration of the agreed upon data

sources, SAS will build analytics that will allow the State to focus on detection of key

fraud behaviors.

The following list includes several sample areas of focus that can form the baseline

of the detection capability SAS can provide. As this is NOT an all-inclusive list, SAS

will work with the State to refine and prioritize the scope of the detection capability.

Misreporting/misclassification of employees: employer, owner,

policy/premium payment, and employee information will be linked together to

provide a holistic view of employer behavior

• Evaluation of premiums paid over time to detect anomalous behavior

• Detection of employee transfer fraud—transfer employees to new

businesses with different profiles and cleaner claims history to obtain

better rates

• Evaluation of claims/injury types in relation to employee classification

• Leverage audit referral data—history of multiple referrals across

single or linked entities drives risk

Underreporting of employees/hours

• Evaluation of claims frequency in relation to employees/hours

reported

• Evaluation of IRS, licensing, department of revenue, and permit data

to determine if it is in line with workers/hours reported (uncover late

or under reporting)

• Evaluation of historic premium payments (receivables)—difficulties

can be a leading indicator of financial stability that can lead to

underreporting

Detection of an underground economy (unregistered businesses/

subsidiaries)



• Linking of cross agency data (e.g., IRS, department of revenue) to

determine if businesses exist but are not paying workers’

compensation

• Evaluation of wage claims against unregistered businesses

• Evaluation of claims filings for unregistered businesses

• Evaluation of structured and unstructured data from field audits and

inspections

Workers’ Compensation Claims Fraud

In addition to the data sources listed in the Workers’ Compensation Premium

Evasion section, additional data is recommended for enhanced detection of workers’’

compensation claimant and provider fraud.

Workers’ compensation claims (line item)—all specific claim line items,

inclusive of information for all injuries, diagnosis, and medical procedures

performed

Claim payments—payment transactions to the individuals filing claims

Medical bill payments—line item payment transactions to medical providers

for services rendered

Registered medical service providers—medical providers and all static

details (e.g., contact information, ownership structure, ownership contact

information) that are registered with the State or as preferred providers

Medical referral information—doctor referral information that is tracked as

part of the medical procedure data

Fraud case details—outcomes and information from past fraud investigations

on claimants and providers

Known bad lists—hot lists of individuals and entities that have been flagged

for past fraud

Through leveraging these additional data sources, SAS will build analytics that will

allow the State to focus on detection of both employee claim fraud and provider claim

fraud. Listed below are several sample areas of focus that can form the baseline of

the detection capability SAS can provide. As this is NOT an all-inclusive list, SAS will

work with the State to refine and prioritize the scope of the detection capability

Workers’ compensation claim fraud (employee and employer fraud)—

claimant, employee, medical provider, and payment information will be

automatically linked together to provide a holistic view of claimant behavior.

• Evaluation of claims paid over time to detect anomalous behavior

(e.g., multiple claim activity)

• Evaluation of claim timing (e.g., seasonal claims, after long layoff)

• Detection of claimant/injury malingering (e.g., time based, procedure

based, working while injured)

• Evaluation of discrepancies between accounts of the incident

(employer disputes, injured on personal time)

Ferreting out fraud



• Evaluation of claims/injury types in relation to incidents and

employee job function

• Evaluation of claims occurring immediately after policy

inception/changes (e.g., reporting additional employees) or in

relation to major employer events (e.g., layoffs)

• Assessment of misalignment of employment details with claim (e.g.,

timing of claim, income not aligned with employment)

• Evaluation of treatments (e.g., timing, diagnosis to procedure

alignment)

• Assessment of suspicious engagement of legal counsel

Workers’ compensation provider fraud—claimant, employee, medical

provider, and payment information will be automatically linked together to

provide a holistic view of provider behavior.

• Evaluation of line item claim data to detect abnormal behavior

o Misalignment of procedure performed

o Excessive treatment or discrepancies

o Inconsistencies in medical billing information

• Evaluation of claims paid and procedures performed over time to

detect anomalous behavior

o Reputation for overbilling / miscoding / unbundling

o Submission of fictitious claims

o Fictitious providers (shell companies)

• Detection of invalid or inter-organization referrals

• Evaluation of consistent medical providers and attorneys involved in

high risk /suspicious claims

3. Advanced Analytics

• Clustering/Segmentation

• Data Mining

• Predictive modeling

• Advanced Statistical Analysis

• Cause-Effect Validation

Advanced Analytics is the core service provided by the SAS team using the SAS

Fraud Framework. Some of the services offered by SAS consultants include: Customize fraud scoring models and investigator alert management to adapt

to the State’s requirements

Apply predictive modeling by taking known fraudulent claims or behavior and

overlay that on new aims data being processed Network data sources:

• Provides enhanced network collusion detection and search capability

based upon association to known fraud

• Improves fraud handling efficiency through data aggregation and

visualization tools and reduction of false positives

Owning 35.2 percent of

the market, SAS

advanced analytics

software dwarfed sales

of the next nine vendors

combined, who together

held only 22.9 percent

of the 2010 market



Employ SAS’ Social Network Analysis to link entities together and investigate

fraud in a meaningful way. By finding and establishing those linkages

between the entities, our team essentially creates a single rollup that is then

subjected to rules, anomaly detection, and predictive modeling. This is a

critical component to the SAS hybrid approach as it allows us much needed

insight into increasingly sophisticated fraud rings.

Enhance business rules, anomaly detection, and predictive models added to

the network analysis for broader protection:

• Deploy anomaly detection models for specific areas of concern

• Build relevant customer state vectors and clusters for predicting out-of-

form behavior

• Improve fraud handling efficiency through data aggregation and

visualization tools and reduction of false positives

4. Build Data Access Skills and Understanding (Foundational) Training

• Generic and/or tools specific

SAS is dedicated to meeting the diverse needs of our customers and strives to make

every training event a successful one. SAS began its software training program in

1976 with the first course given to the U.S. Department of Agriculture. Since then,

SAS has trained thousands of state and federal government employees. SAS’

comprehensive training curriculum consists of more than 200 courses that address

all aspects of SAS software. SAS offers courses for users of all experience levels,

from new computer users to seasoned professionals. As the developers of SAS

software, we provide the most up-to-date training available on the SAS System.

Instructor-based training courses allow students the opportunity to learn the course

material in a desirable time frame. SAS has a training centered located in

Minneapolis, Minnesota with two state-of-the-art classrooms and local instructors.

We combine lectures with textbook exercises and computer workshops to reinforce

each lesson taught.

C. A description of any tools, hardware, and software that would be needed to

support the services you are offering for the subject matter;

What distinguishes SAS is our ability to complete the project with expediency and

low risk, while delivering value early in the project lifecycle. We are also able to

immediately focus on the areas that will result in the significant cost avoidance

benefits that are critical to the State especially during these difficult economic times.

For this project SAS will be accountable as the prime contractor to provide complete

software, hardware, services, training, maintenance and support in one fully

integrated solution hosted by SAS Solutions OnDemand. The sole objectives of

SAS Solutions OnDemand, formed in 2000, are to provide customers with rapid

deployment, best practices in the use of SAS solutions, scalable and highly secure

hosting environment, all wrapped in a full service approach. This approach provides

Comprehensive training

curriculum geared for all

experience levels

Pinpoint focus on the

areas of most benefit

SAS Solutions

OnDemand offers rapid

deployment



the State with a one-stop shop to answer any questions whether solution related,

infrastructure related, or any facet of the users everyday working environment.

For each OnDemand customer, a professional team is assigned to work with the

customer’s specific solution and support a wide variety of questions, additional

requirements, issues, solution enhancements—all directly connected to SAS’

research and development. SAS Solutions OnDemand’s is committed to on our

customers obtaining the most value from SAS solutions as quickly as possible.

Through this approach, SAS has redefined the concept of hosting services with a

holistic approach that integrates three interwoven components into a unified offering:

Infrastructure– A secure, high-performance data processing infrastructure,

with 99% uptime guaranteed in service level agreements. Additional service

levels are available as additional options if required by the client.

Expertise—“The right expert at the right time” for optimizing the

infrastructure, the data warehousing foundation and the business intelligence

applications = the total solution.

Communication—A single point of contact for customer liaison and project

management, with “the buck stops here” accountability for the end-to-end

solution.

Customers enjoy the benefits of a comprehensive solution based on the industry’s

leading analytics; delivered by the insiders who know these tools best; and deployed

on an infrastructure that has been tuned specifically for the solution at hand. SAS

Solutions OnDemand delivers actionable intelligence about workers’ compensation

fraud, but also solid customer relationships and organizational performance.

The SAS Advanced Analytics Lab team is a group of highly proficient statisticians

and modelers within the SAS Solutions OnDemand department. Using SAS software,

this team developed the innovative analytical processes and techniques inherent to

the SAS Fraud Framework Solution. This team has delivered end-to-end solutions

from raw data as well as advanced analytics solutions encompassing data

warehousing, data quality, and documentation as part of our Solution Delivery

Methodology. The Advanced Analytics Lab has the flexibility and competence to

address emerging and challenging analytic issues within all business sectors.

SAS customers receive a full suite of support services at no extra charge, including

skilled telephone technical support and unlimited, around-the-clock online technical

support. Our online customer support center provides always-on access to a wealth

of technical support, reference information, educational resources and communities.

Knowledge-sharing is continuously available through regular seminars, Webcasts

and an expansive selection of training courses. SAS OnDemand customers are also

provided with a web-based issue-tracking and communication tool with severity alerts

Industry-ready solutions

for fast “time to value”

Over 140 government

and commercial

customers entrust their

data and applications to

SAS Solutions

OnDemand

SAS Solutions

OnDemand offers the

complete solution—

easing the burden on

the State’s IT resources

SAS has one of the

highest concentrations

of PHDs of any other

entity across the country



escalating based on issue categorization, as well as an integration on-line project

documentation solution.

The SAS Fraud Framework for Worker’s Compensation The opportunistic fraudsters of yesterday, who typically exploited a single

government service or program, have evolved into sophisticated and organized

networks attacking every available program and service while moving at light speed.

The tax dollars lost to fraud are increasing at an alarming rate. Although most

workers' compensation claims are perfectly legitimate, the FBI estimates that 10

percent of all claims contain some element of fraud.

Traditional program or agency controls are no longer sufficient to keep up with

today’s fraudster. Governments must look to a more holistic and enterprise level

approach that spans program and agency boundaries to more effectively track and

identify both opportunistic fraudsters and sophisticated and organized fraud rings and

criminal networks.

The SAS Fraud Framework is enterprise in scope, allowing the State to create a

holistic view of entities across all government programs and services, and aggregate

their activity in such a manner as to detect discrepancies across various government

touch points that could be easily missed when only looking at one program or service

area. In addition, it allows activities to be easily consolidated at the entity level to

streamline investigations, recoveries and, if necessary, criminal prosecution. Thus,

access to pertinent information is available to any internal or external resource

having the appropriate privileges—providing a way to effectively analyze the wealth

of available data and turning it into actionable information. In addition, the solution

supports a wide variety of fraud types and exposures, including worker’s

compensation and unemployment insurance, tax, healthcare programs such as

Medicaid, and other social services programs such as TANF, WIC, Child Care, etc.

As shown in Figure 4.6.1, the solution is built on the SAS Business Analytics

Framework, a complete end-to-end data integration, business intelligence, advanced

analytics and business solution framework. The SAS Fraud Framework supports the

entire end-to-end fraud detection process, including data integration, analysis, alert

generation, investigation and ultimate disposition—be it an audit, recovery action,

criminal prosecution or decision to take no action. Because the solution helps to

manage the entire fraud discovery and investigation workflow, pertinent information

flows seamlessly throughout the process, and the outcome of each and every

investigation is fed back into the detection engine to fine tune the analytics over time.

Fraud increasing at an

alarming rate demands

solutions keep up

Enterprise approach to

fraud detection =

thought leaderships

The SAS Fraud

Framework addresses

all types of fraud

End-to-end fraud

detection process



Figure 4.6.1. The SAS Fraud Framework solution is built on the SAS Business Analytics Framework, a complete end-to-end data integration, business intelligence, advanced analytics and business solution framework. The SAS Fraud Framework supports the entire end-to-end fraud detection process, including data integration, analysis, alert generation, investigation and ultimate disposition. Prevention and detection are the two key gears guiding the solution.

Since fraudsters are becoming increasingly more sophisticated in their ability to

detect and circumvent traditional control measures, the SAS solution leverages a

hybrid approach to fraud detection. Rather than relying on a single detection

technique, the solution combines heuristic rules to find known fraud patterns,

anomaly detection to surface unknown yet unusual behavior, and predictive models

to allow the system to leverage attributes of past fraud cases to identify future cases

of fraud within the State’s data. And finally, the solution incorporates sophisticated

link analysis and social networking capabilities to detect hidden relationships

between entities—detecting larger fraud networks and collusion. This hybrid

approach allows the system to cast a very wide net to identify a wide variety of both

known and unknown fraud schemes, and then score, rank and prioritize suspect

entities in such a manner as to enable a limited staff of investigators and auditors to

concentrate on the most important and highest risk entities for investigation.

SAS recently earned the Technology Innovation of the Year Award in the field of

enterprise fraud detection and prevention by Frost & Sullivan for its fraud

management solution. After thorough analysis of the fraud solution market, the

research firm found that SAS Fraud Management led the industry in integrating

The SAS Fraud

Framework can help

you stay one step

ahead of today’s

fraudsters



advanced analytics, superior decision management capabilities and sophisticated

rules into a complete scoring and reporting platform. SAS has consistently been

positioned in the Leaders Quadrant of the Gartner Magic Quadrant for products such

as data quality, enterprise business intelligence and data mining. Not only is SAS the

largest vendor in the overall data mining market, we have the most established track

record of customer successes. Gartner cites our strength being due to our wide set

of tool-based capabilities along with our expanding numbers of packaged solutions.

SAS remains the overwhelming leader in the advanced analytics market, according

to IDC, with 33.2% market share in 2008 (IDC, June 2009). Voting with their IT

budgets, companies and government chose SAS Analytics more often than the next

16 analytics suppliers listed COMBINED, according to June09 Worldwide Business

Intelligence Tools 2008 Vendor Shares report.

SAS Fraud Framework Overview

The fraud detection capabilities of the SAS Fraud Framework are delivered through a

range of innovative approaches driven by the needs of SAS customers across

industries. The sophisticated data integration, analytics, and fraud investigation

platform is proven to scale by amount of data processed, quantity of users, and

number of business areas addressed. This scalability will allow the State to not only

expand the capability within the SAS Fraud Framework, but expand to other fraud

detection areas to resolve subsequent business issues.

The advanced analytic approaches and fraud risk scoring techniques, along with the

powerful entity linking and social network building capabilities, create a complete

intelligence picture of all of the entities and risks associated with a case. From

opportunistic fraud detection, to locating sophisticated and organized fraud that is

invisible when using traditional techniques, the SAS Fraud Framework will

proactively produce fraud alerts/referrals earlier in the cycle and then support the full

investigation process—ultimately finding more fraud, faster, and greatly reducing

losses and exposures.

As the laws and environment of each state vary, we have found pre-built models are

useful, but not as valuable as customized models for identifying fraudulent behavior.

Thus, the SAS Fraud Framework has been built on a foundation of flexibility in the

data structure and detection engine, allowing modification and customization of

analytics models while minimizing the effort required. In order to detect the State’s

fraudsters, our proven approach for detecting fraud includes customizable market-

specific business rules, predictive models, anomaly detection, and social network

analysis based on the State’s unique data. In addition, our implementation team has

valuable experience in detecting these types of fraud which they bring to bear rather

than relying solely on canned or pre-built models. Members of the SAS Advanced

Analytics Lab will leverage our vast experience in workers’ compensation and

unemployment insurance fraud detection to apply proven analytical models, methods

Scalable platform to

expand to areas outside

of workers’

compensation and

unemployment

insurance

SAS’ hybrid approach to

fraud proactively detects

fraud and gives

complete intelligence

picture

The solution is both

automated and

customizable

Advanced Analytics Lab

tunes models based on

their experience

Owning 35.2 percent of

the market, SAS

advanced analytics

software dwarfed sales

of the next nine vendors

combined, who together

held only 22.9 percent

of the 2010 market



and templates to the State’s unique data, thus providing a solution that is based on

industry best practices but precisely tuned for the unique circumstances within the

State.

From SAS’ advanced hybrid approach to fraud analytics to the integrated end-to-end

fraud prevention solution, to the intuitive user interface, the SAS Fraud Framework

offers a wide array of competitive advantages. It is only with SAS that the State will

be able to:

Proactively prevent more fraud through a holistic, enterprise view of the big

picture

• Automated associations of related perpetrators and a clear

understanding of all behaviors

• Aggregation of fraud alerts and cases to arm investigators with all

pertinent information

• Better understanding of emerging threats so action can be taken

before substantial losses occur

Effectively prioritize alerts and cases for fraud investigators using a

configurable value based approach

Significantly reduce false positives through the application of the hybrid

approach and analytical based scoring

Maximize investigator efficiency through the automated aggregation of data

and the streamlined investigative tools and interface

Drive consistency into the end-to-end fraud prevention process

• Consistency in fraud alert generation through data driven detection

• Consistency in investigation process to ensure minimum standards

in case file output for ready consumption by law enforcement

Reduce overall total cost of ownership and be able to rapidly implement

fraud detection in new business areas by leveraging a single enterprise-

wide, flexible and extensible, end-to-end fraud detection platform

The SAS Fraud Framework process consists of a series of components that support

end-to-end fraud detection. Included in the SAS Fraud Framework software package

is all applicable software components needed for combating fraud, including:

Data access, quality and integration

Detection and alert generation (including fraud rings and networks)

Analytic model generation tools

Enterprise business intelligence and reporting

Alert management and investigation

Enterprise case management

As depicted in Figure 4.6.2., the SAS solution facilitates the entire end-to-end

process for fraud detection and prevention. The following sections outline how the

SAS Fraud Framework will work for the State:

Differentiation from the

competition

All components needed

to address fraud



Figure 4.6.2: High-level SAS Fraud Framework process flow shows how the solution ingests data for exploratory data analysis and transformation, stages it, merges business rules and then applies advanced analytics such as anomaly detection, predictive modeling and social network analysis to generate alerts. The alerts are then surfaced in an easy to understand interface and can be integrated with an existing or SAS’ case management solution. The solution has a “learn and improve” cycle in order to fine-tune results over time as more information is added to the system.

Data Access, Quality and Integration

SAS enterprise data integration is a powerful, user configurable and comprehensive

component of the SAS Fraud Framework that can access virtually all data sources;

extract, cleanse, transform, conform, aggregate, load and manage data; support data

warehousing, migration, synchronization and federation initiatives; and create real-

time data integration services in support of service-oriented architectures. The

power behind SAS data integration is that it will support the integration and

configuration of disparate data sources to create a comprehensive person profile

across various agencies and source systems.

The SAS Fraud Framework leverages SAS enterprise data integration to draw in

data from all relevant sources, including cross-agency and third-party, and parses

the data to create a complete data model of all entities and their respective attributes

and activities. Identity and relationship resolution capabilities are employed to

consolidate data when multiple source records exist, creating a single view of each

entity by applying sophisticated data matching, precedence and lineage rules to

resolve inconsistent or ambiguous data. This aspect of data quality is especially

Access virtually any

data source

Sharing data across the

State –removing

traditional barriers



important in fraud detection, since persons with intentions of cheating the system will

often provide inaccurate or conflicting information. As these differences across

records and systems can often be a warning sign of fraud or other improper

activities, the system maintains all original values, while leveraging precedence rules

developed in concert with each customer to establish what is the most trusted value

to use for initial display purposes within the application—ultimately providing a clear

and concise composite depiction of each entity.

Because of its ability to integrate data from any source, the solution provides the

capability to create a comprehensive person profile (whether natural or juridical) that

is then available to all users of the system provided they have the appropriate

security privileges. The sophisticated matching engine provides both deterministic

and probabilistic matching capabilities that allow match sensitivities or scores to be

set at the individual attribute level. Match codes can be assigned and persisted on

an individual record basis, providing an extremely efficient matching process.

An example of this capability is State of North Carolina Department of Justice, who

uses SAS to consolidate information across all criminal justice entities to create an

offender centric view of all offender information. Data sources span multiple

branches and agencies of state government, local courts, jails and arrest data, and

select federal information. The system provides robust search and watch capabilities

so that any criminal justice professional can get complete, integrated, accurate and

timely offender information at the touch of a button. Within the context of this

application, SAS helped the North Carolina address a number of challenges, both

technical and cultural. We assisted the state with developing data governance

standards and process for inter-agency data sharing to ensure that sensitive data

was adequately secured and protected. In addition, we helped the state develop

data precedence rules to establish which data elements were considered to be either

the most trusted or the “value of record” when conflicts existed between records.

The metadata driven approach of SAS enterprise data integration enables the SAS

Fraud Framework to ingest data in virtually any format, including unstructured free

text. Because the solution leverages SAS’ award winning data integration tools, the

underlying data model and data integration process can be easily maintained,

enhanced and extended. This allows the system to maintain a comprehensive

dossier on entities that includes consolidated identity and descriptive information,

transactions, changes and associations across multiple agencies and systems and

extending even into external data sources. The tools provide an easy to use

graphical interface for managing the data integration process, allowing new data

sources to be added as they are identified and available.

Create a

comprehensive person

or citizen profile with the

SAS Fraud Framework

data model

SAS is helping to

ensure criminals in NC

do not fall through the

cracks

The SAS Fraud

Framework is built to be

shared and extended



Detection Engine/Alert Generation Process

The Alert Generation Process is an advanced detection engine that applies business

rules, anomaly detection algorithms, predictive models, and social network analytics

on the ingested data to raise alerts on entities (e.g., claims, participants, employers)

and networks for fraud. The Alert Generation Process aggregates scores generated

by the analytics at the appropriate entity level The Alert Generation Process uses

these aggregated scores to prioritize the alerts for consumption.

Produces independent and combined scores to assess overall risk on an

entity and network basis, and then generates alerts that are automatically

routed to the appropriate resources.

Scores entities and their full histories—scores associated networks using

behavioral data (patterns, network growth rates, activity levels) and other

data provided (current/previous addresses, contact numbers, employers).

Performs batch processing of existing participants, claims, and networks,

enabling you to detect and investigate existing fraud as well as prevent new

fraud.

Ingests existing risk scores or red flags and provides an aggregated score at

the network level to identify subtle or hidden relationships.

Can be executed as a scheduled batch job or run on an ad-hoc basis.

One key differentiator of the SAS Fraud Framework that drives incremental fraud

detection earlier with reduced false positive referrals is the SAS hybrid approach to

fraud analytics. The SAS hybrid approach contains business rules, anomaly

detection algorithms, predictive models, and Social Network Analytics to produce

targeting schemes. It then utilizes sophisticated scoring and ranking to surface the

most relevant referrals for investigation. The content and logic within the four pillars

of the hybrid approach are specific to SAS Fraud Framework and are configurable to

the State’s data and specific industry sectors and fraud detection objectives. That is

to say, SAS has sets of rules for detecting certain types of fraud (workers’

compensation, medical claims fraud, etc.) that can be used “off-the-shelf.” The

hybrid approach augments these rules with additional tools that are built specifically

for the climate at hand.

As depicted in Figure 4.6.3, the SAS hybrid approach leverages the following

techniques:

Business Rules—encode known fraud schemes and indicators and are a

good first measure for identifying suspicious activity. Unfortunately, rules

alone are easy for sophisticated fraudsters to spot and avoid by managing

their behaviors under the rule threshold levels. However, rules in

combination with anomaly detection, predictive models, and social network

analysis provide a strong capability to detect the fraud that is acting below

SAS’ hybrid approach

reduces false positives

Aggregate scores to

prioritize alerts

Unique to SAS, these

techniques ensure no

fraudsters can fall

through the cracks



the rule thresholds. The SAS Fraud Framework includes built-in rules for

common types of fraud.

Figure 4.6.3. The SAS hybrid approach to fraud detection includes the integration of multiple disparate data sources and first adding business rules to filter fraudulent behaviors. Anomaly detection then detects individual and aggregated abnormal patterns. Predictive models are created to assess against known fraud cases. Finally, social network analysis adds knowledge discovery through associative link analysis and can help identify organized fraud rings. This hybrid approach proactively applies a combination of all four analysis approaches at the entity and network levels and ensures fraud does not slip through the cracks.

Anomaly Detection—compares entity behaviors and attributes to peer

groups or past behavior patterns to highlight activity that is outside the norm

and potentially indicative of fraud. (An entity can be a single transaction, an

individual, or even a network.) Also known as unsupervised methods, SAS

Fraud Framework anomaly detection leverages many analytical techniques,

including, mean, standard deviation, robust regression, clustering, sequence

analysis, and peer group analysis to name a few. These techniques are

critical to uncovering unknown fraud schemes and keeping up with the

changing trends in fraud. The SAS Fraud Framework has existing anomaly

detection algorithms built from SAS’ industry leading advanced analytics

capability in the government space. These anomaly detection algorithms run

in conjunction with the rules at the entity and network levels and highlight

suspicious behaviors to the investigators to assist in investigative efficiency.

Identifies outliers that

may prove fraudulent –

unique to SAS at the

network level



Predictive Models—leverage known fraud case outcomes (known

fraudulent claims, participants, networks, and schemes) to generate fraud

scenario profiles. These fraud scenario profiles are used in the review of the

enterprise data to uncover similar behavioral patterns and raise alerts on

entities and networks that are demonstrating similar fraud attributes as the

known cases. Also known as supervised methods, SAS Fraud Framework

predictive models leverage many analytical techniques, including neural

networks, decision trees, generalized linear models, econometric models,

and gradient boosting to name a few. The SAS Fraud Framework has

existing predictive model templates built from SAS’ industry leading

advanced analytics capability in the government space that are configurable

to an organizations data and requirements. These predictive models are run

in conjunction with the rules and anomaly detection algorithms at the entity

and network levels. The application of these predictive models at the

network level is unique to the SAS Fraud Framework.

Social Network Analysis (SNA)—fraudsters often recycle information and

behaviors to spread their activity across identities and programs to obtain the

maximum financial gain without being detected. In addition, collusive

networks of employers, providers and beneficiaries often act in concert to

thwart internal control mechanisms. By linking all data, including claim,

provider, employer, and participant data, the SAS Fraud Framework can

uncover these hidden relationships. Linking, however, is not enough—SAS

has a patent pending approach, Net-CHAID, to refine these links into

meaningful networks for the application of fraud analytics. Once these

networks are created, then rules, anomaly detection and predictive models

are applied at the network level. It is through the application of these

techniques on aggregated data that the SAS Fraud Framework can detect

the most sophisticated and costly fraud rings, with reduced false positives.

The advanced analytic approaches and fraud risk scoring techniques, along with the

powerful entity linking and social network building capabilities, create a complete

intelligence picture of all of the entities and risks associated with an entity and its

related fraud score. From opportunistic fraud detection to locating sophisticated and

organized fraud that is invisible when using traditional techniques, the SAS Fraud

Framework will proactively produce employer fraud scores, related targeting lists,

and audit requests earlier in the cycle and then fully support the entire investigation

process.

Social Network Analysis

Social Network Analysis uses static and transactional data to build links between

entities and uncover the hidden relationships. The automatically generated social

networks (groups of entities linked based upon their relationships with each other)

Proactively produce

employer fraud scores,

related targeting lists,

and audit requests

Uses known fraud to

identify patterns and

create rules

SAS’ Net-Chaid refines

Social Network Analysis

links to identify

meaningful networks.

In fact, it is only with

SAS that organized

crime can be

detected—in addition to

the other types of fraud

Additional layer of fraud

protection uncovers

hidden relationships



are used in the Alert Generation Process for the application of network based rules

and analytical models, providing an additional layer of fraud detection.

The generation of social networks is driven by a metadata configuration. All data

records ingested are linked exhaustively based on combinations of attributes and

behavioral patterns within the data. Then, using statistical techniques, including the

patent pending Net-CHAID algorithm, common entities are identified and collapsed to

produce single views of entities and the links are refined to produce meaningful

groups—or networks—for analysis.

The Alert Generation Process generates scores at the entity level that are then

aggregated during the network build process and used to further refine the networks

based upon common behavioral patterns. Once the networks are fully refined, the

Alert Generation Process applies an additional set of network based rules, anomaly

detection algorithms and predictive models to the social networks to generate

network level scores and alerts. The Alert Generation Process also prioritizes the

alerts for consumption.

Alert Management & Investigation

Alert Management integrates and manages the alerts created in the Alert Generation

Process, providing an interface to perform list review, analysis and investigation on

the output of the Alert Generation Process. The prioritization and queuing of alerts is

configurable to an organization’s requirements, and multiple separate user queues

can be defined and alerts routed to specific fraud triage resources or fraud

investigators based upon an organization’s business rules. Through a Flex based

graphical user interface, Alert Management provides detailed line item fraud alert

information, corresponding source data, analytic scoring results, and the social

network views that are required to support alert triage/disposition and the fraud

investigation process. In addition, alerts from other existing fraud systems can be

consolidated within the Alert Management component to provide a true enterprise

wide view of fraud.

The SAS Fraud Framework utilizes the SAS business intelligence platform for

management of business intelligence and reporting. SAS business intelligence

platform is a comprehensive, easy-to-use solution that integrates the power of SAS

analytics and data integration to share insights that power better business decisions.

It includes role-based, self-service interfaces for all types of users within a well-

defined IT governance framework and a centralized point of administration. This

helps organizations simplify and speed business intelligence deployment.

The SAS Fraud Framework end-user interface provides the capability to create a

wide array of customized reports and dashboards, as well as select pre-defined

reports from a gallery via SAS’ web reporting capability. Pre-defined layouts and

More advanced than

entity analytics

Patent-pending

techniques applied to

data

The SAS Fraud

Framework designed for

all user types

Combine all alerts in

Alert Management for

real view of fraud



templates are also available to ease the report creation process. Output from reports

and dashboards can be integrated with MS Office technologies and exported to

multiple different formats.

Enterprise Case Management

SAS Enterprise Case Management (ECM) enforces best practices and proper

gathering of evidence to complete the fraud investigation process. ECM provides a

structured environment for managing investigation workflows, attaching comments or

documentation, and recording financial information. Case management is the

backbone of documenting investigation processes, exposures, and losses and of

driving consistency into the process. The user defined requirements to drive

consistency will help ensure that case files are completed appropriately and that

cases referred to law enforcement and the district attorney have sufficient evidence

for prosecution. Additionally, ECM is expandable to include prosecutors and law

enforcement groups as end users.

ECM is a configurable work flow based case management solution with a full

document repository for storing and tagging investigation and case documentation.

ECM receives referrals and incident alerts requiring investigation from multiple

monitoring systems, including Alert Management, and is utilized to support and drive

consistency into the field investigation process and case file creation. The automatic

pre-population of field level data from Alert Management streamlines the

investigation process and prevents rekeying errors. The disposition and tracking of

case outcomes is a critical element of ECM that supports the enhancement of future

monitoring and overall operational efficiencies based upon historical information. In

addition, the ECM solution can import existing case records and historical information

from existing tools to ensure a full auditable history is maintained.

The SAS Fraud Framework can also integrate with existing case management

utilities.

Learn and Improve Cycle—Intelligent Fraud Repository

Each time an alert is actioned or a fraud referral/case completed, the results are

stored within the Intelligent Fraud Repository as known outcomes. The predictive

models used in the Alert Generation Process access this repository of known

outcomes in the application of the supervised analytic approaches for the SAS Fraud

Framework.

This feedback loop allows the SAS Fraud Framework to continue to learn from

investigative results and further improve precision by raising more alerts on entities

and networks that have similar attributes to confirmed fraud cases. In addition, this

supports the SAS Fraud Framework in keeping up with the changing trends in fraud,

by integrating newly confirmed fraud results into the Alert Generation Process.

Structured environment

for managing

investigations

ECM makes

investigator’s jobs easier

The SAS Fraud

Framework never stops

learning and evolving

Known outcomes feed

the models



Alert Administration

Alert Administration allows system administrators to administer and maintain the

business rules, models, and network analytics run within the Alert Generation

Process. Through the GUI of the Scenario Administrator, Fraud Business/Process

Administrators can easily configure the details of the rules and models and the batch

frequencies without support from IT. Through this interface, administrators can

configure various populations (industry segments, geographic regions, etc.) to target,

and develop specific detection schemes or thresholds for each defined population.

SAS Fraud Framework’s data mining is used to generate and test the validity of new

fraud business rules and analytical models. Once validated, the new rules/models

can be readily deployed from SAS Fraud Framework data mining to the SAS Fraud

Framework and managed via Alert Administration. SAS Model Manager provides an

easy-to-use graphical user interface that guides fraud analysts through a repeatable

process for registering, testing and validating models. Model Manager also supports

the champion vs. challenger capability of confirming the best fit models.

D. To the extent you are able, indicate the level of State involvement necessary

for the successful implementation of the services you are offering. Include

estimates for resources levels and time.

The level of State involvement will vary from project to project. We have included a

high level roles and responsibilities chart in Section E for your information.

E. A list of personnel anticipated to provide professional services under this

master contract program; provide a narrative of their individual qualifications

or provide a resume or curriculum vitae specific to each proposed subject

matter category;

The following table shows typical SAS roles and responsibilities:

SAS Role Typical Responsibilities

Project Director, Advanced Analytics Lab

Provide overall program leadership for development, strategy and direction of the SAS

Fraud Framework.

Facilitate cross-departmental activities and resources within SAS in order to ensure

success of the State project.

Analytics Manager, Advanced Analytics Lab

Review, assess and provide recommendations for business and functional

requirements.

Supervise State Analytic Consultants and provide guidance with respect to data

mining, text mining, Social Network Analysis, forecasting and analytic model

management.

IT Manager – Advanced Analytics Lab

Review, assess and provide recommendations for State system and infrastructure

requirements.

Supervise State implementation and provide guidance with respect to SAS Solutions

OnDemand best practices, processes and IT policies. Quality Review and approve all State Quality Assurance test plans.

Provides easy

configuration and

distribution of alerts

Confirming and

validating the best

models to include in the

solution



Assurance Manager -Advanced Analytics Lab

Review and approve all State Quality Assurance test results.

Review and approve State quality control plans.

Review and approve State quality management deliverables.

Ensure adherence to SAS methodology processes and best practices.

Assign State testing responsibilities when Quality Assurance is needed.

Create project quality plans (Integration Test Plan, System Test Plan, and Quality Plan

Worksheet) for the State project.

Coordinate acceptance of State code subsystems and programs with appropriate staff.

ETL Development Lead, Advanced Analytics Lab

Build, implement and manage data integration processes for State data sources.

Develop State metadata environment.

Perform impact analysis of potential changes made across all State data integration

processes.

The proposed solution shall provide interactive debugging and testing of jobs during

development and give full access to logs.

Develop capabilities to connect to all required State data sources.

Build SQL functionality to load State data.

Fraud Analyst Lead and Application Developer, Advanced Analytics Lab

Develop and implement data mining features for interactive models such as rules

building and exploratory data analysis capabilities.

Develop and implement social network analysis functionality such as alert generation,

network level scoring and querying using advanced analytic and statistical methods

such as associative cluster analysis.

Develop and implement text mining tools to provide comprehensive text preprocessing,

multi-format document support and model score code generation.

Create forecast modeling tools.

Create analytic model management tools.

Project Manager

Project management and administrative responsibilities of the State project.

Coordinate data mining, text mining, SNA, forecasting and analytic model management

development.

Work with State users to identify all system level, data, business, technical and

functional requirements for the State project.

The following table shows the State’s typical roles and responsibilities:

Role Responsibilities Analytic Engineer (optional)

Design analytic applications (only if the State plans to design its own analytical work).

Business Analyst

Advise the project team on the key industry business objectives and the definition of

the business process.

Advise the project team on specific subject matter, including industry trends, and

business systems requirements.

Help identify and evaluate the business value of the project. This may involve

establishing and measuring ROI.

Communicate high-level business requirements to team.



Participate in requirements meetings.

Provide information on business needs and existing business processes and required

enhancements to those processes.

Provide analytical and reporting requirements.

Review design to verify that it meets requirements.

IT/Data Administrator

Address security issues.

Administer user access.

Compliance / user access lead is primary contact of SIRT team if a security incident

occurs.

Advise project team on the State’s overall IT strategy, processes, and policies.

Coordinate the IT staff to support the project team.

Manage the data extract process.

Ensure consistency and integrity of data.

The data stewards have read/write access to the metadata, allowing them to perform

the following functions on tables within the metadata environment: read, write, alter,

and create.

Data Architect/Data Modeler

Perform logical database design.

Provide information on the design of the operational data sources and responsible for

reviewing the subject data models, dimensional data models, and physical data

models.

End-user or Business Users

Advise project team on business requirements, and desired attributes.

Actively provide feedback on intermediate deliveries, if any.

Assume responsibility for the quality of the system, and perform User Acceptance

Testing

The end-users can include

• Business users

• Business unit analysts

• User representatives

• Test users

• Business unit managers.

Develop User Acceptance Test Plans.

Conduct User Acceptance Testing.

Provide results of User Acceptance Testing.

Executive Sponsor

Provide overall direction for team.

Promote cross-departmental activities.

Facilitate resolution of cross-departmental issues.

Create environment conducive to teamwork.

Ensure adequate funding is available.

Manage expectations of executives.

Identify the key members of the project’s management team.

Establish key priorities.

May authorize and/or finance the project.

Notified by SIRT team if primary contacts cannot be reached.



When beginning the project, SAS assembles the initial project team. The team

usually includes one or more members of the sales/proposal team along with an

allocated account manager or director, a project manager, and a technical lead. At

appropriate points throughout the project as identified in the project plan, the SAS

team may expand to include other resources as applicable.

All nominated SAS staff allocated to any agreed project role possess the required

skills, experience and knowledge required to meet assigned expected work products.

SAS strives to hire and retain the best employees in the industry. Many SAS

consultants hold bachelors, masters, and doctoral degrees in such areas as

computer science, statistics, operations research, and business administration. SAS

consultants are also are experienced in detailed consulting operations, applications

development, system analysis and design, and project management.

When possible, SAS attempts builds teams using consultants who have previously

worked together or who share specific domain knowledge relevant to the project

expected work products. This further extends the benefits of the SAS team

approach. The following profiles represent examples of SAS employed individuals

who could be made available to assist with the proposed project or engagement.

Specific consultants allocated to any project are assigned based on the actual

requirements of the proposed project and availability at the time of contract

commencement.

Project Director—(J.B.)

Relevant Experience

SAS Institute, Inc. (1981 – Present)

Vice President, SAS Solutions OnDemand (2008 – Present)

Product research and development, marketing, sales, program management,

IT, quality assurance, and documentation. Current products and services

include SAS Solutions OnDemand, SAS Fraud Framework, Advanced

Analytics Lab, SAS Drug Development, Education Value Added Assessment

System, ASP educational testing.

Senior Research and Development Director (2000 – 2008)

Product research and development for the Analytic ASP within Research and

Development.

Research and Development Director (1999 – 2000)

Product research, development, and marketing for Analytic Solution Products

within the Business Solution Division.

Product Development Director, Data Mining (1997 – 1999)

SAS provides the right

skills needed based on

the requirements of the

project

Initial project team

SAS team approach



Product development of Enterprise Miner, Quality Assurance, and product

Marketability.

Manager, Statistical Training and Services, (1988 – 1996)

Support for all statistical training and consulting support.

Manager, Statistical Training (1983 – 1988)

Provided customer software support for all statistical training needs.

Intensive travel around the world and participation in on-going development

of product to customer needs.

Pioneered on-site customer and general public statistical training program.

Statistical Training Specialist (1981 – 1983)

Taught product software lines to limited customer base. Due to the newness

of the company, much time devoted to course development for on-coming

statistical software product lines.

North Carolina State University (1999 – 2002)

Adjunct Professor of Statistics and Genetics

Self-Employed (1981)

Contracted by SAS as an instructor.

Employed into a new department to establish continuity of statistical training

style. After less than three months, employed full time.

Department of Statistics, North Carolina State University (1978 – 1981)

Research Associate

Consultant to U.S. Department of Agriculture and on-going studies to receive

Ph.D.

Department of Statistics, North Carolina State University (1978)

Statistical Instructor

Department of Biostatistics, State University of New York (1977)

Statistical Instructor

Vanderbilt University (1976 – 1977)

Research Associate

Subject Matter Expert—(J.R.)

Relevant Experience

SAS



Solutions Architect, SAS Global Fraud and Financial Crimes Global Practice

Principal for Insurance Fraud Solutions

Responsible for the planning, management and marketing of fraud detection

and investigation management solutions for the property, casualty, life and

disability insurance markets worldwide.

The Hartford Financial Services Group

Director of SIU Strategic Operations for the Property & Casualty Special

Investigations Unit

At one of the nation’s largest insurance and financial services operations,

J.R. established and managed a 20-member analytical and intelligence

operation within the SIU and designed analytical protocols for investigating

auto, property, liability, and workers’ compensation claims.

Hartford Life

Manager in the Group Benefits Division Special Investigations Unit

Supervised a team of Investigative Analysts and developed tools and

resources for investigating suspicious life and disability claims.

LexisNexis

Product and Markets Manager, Fraud Solutions Group

Education and Awards

Master’s degree in Economic Crime Management

Bachelor’s degree in Criminal Justice

Utica College Economic Crime and Fraud Management MBA program,

Adjunct Faculty

Certified Fraud Specialist (CFS)

Certified Fraud Claim Law Specialist (FCLS)

Co-inventor of a patent-pending insurance fraud detection model that utilizes

determinate and indeterminate data

Executive board member of the New England Anti-Fraud Association

(NEAFA)

Trainer for the Association of Certified Fraud Examiners (ACFE),

International Association of Law Enforcement Intelligence Analysts (IALEIA),

National Healthcare Anti-Fraud Association (NHCAA), and the International

Association of Special Investigation Units (IASIU), as well as working groups

for the Federal Bureau of Investigation (FBI) and other law enforcement

agencies.



Subject Matter Expert (C.B.)

Relevant Experience

Forrester & Dick

Consultant 2010 – Present

Consultant providing advice on matters of general liability and workers’

compensation insurance matters as well as consultation to clients on matters

of public policy, insurance and labor laws.

State of Louisiana - Louisiana Workforce Commission 2008 – 2010

Director, Office of Workers’ compensation

Managed staff of 157 and an annual budget of $62 Million

Legislative Lead for Louisiana Workforce Commission

Worked with Agency Executive Team and Governor’s Policy Team to

develop policy priorities.

Drafted legislation and handled committee testimony for agency bills.

Revamped fraud division audit selection and investigative procedures

resulting in a 60% increase in investigations and a 700% increase in referrals

for criminal prosecutions.

Lead in the procurement and implementation of a statewide fraud, waste and

abuse detection platform.

Served as interim executive counsel for Executive Director, Louisiana

Workforce Commission (October, 2009 – November, 2010)

State of Louisiana Property and Casualty Insurance Commission – Member –

2008-2010

The Property and Casualty Insurance Commission reviews and examines

the availability and affordability of property and casualty insurance in the

state of Louisiana and makes policy recommendations to the Louisiana

Legislature.

Chairman of the Governor’s Advisory Council on Workers’ Compensation

(2008-2010)

Committee of 17 stakeholders from business, labor and the medical

community charged with advising the Governor and the Legislature on policy

matters pertaining to workers’ compensation.

Chairman of the Louisiana Workers’ Compensation Second Injury Board (2008-

2010)



Subject Matter Expert–(G.H.)

Relevant Experience

SAS Institute Inc. (September 1997 – Present)

Government Practice Director, Fraud and Financial Crimes Global Practice

2005 – Present

Responsible for field support and product direction in applying SAS’

fraud detection and prevention capabilities within the government

market.

Sr. Systems Engineer May 2004 – 2005

Responsible for technical relationship with flagship State and Local

Government customer.

Duties included developing overall account strategy, developing key

messages and collateral to support sales initiatives, pre-sale customer

meetings and presentations, and post-sale customer care activities.

Grew account revenue by 600% over a 4 year period and developed

customer into a premier partner and reference.

Also acted as strategic advisor to business unit General Manager to

assist with repeating success throughout the business unit.

Anti-Money Laundering Product Manager December 2002 – May 2004

Manage overall product life cycle, including market opportunity

evaluation, requirements definition, go-to-market and launch plans, sales

readiness and release cycles.

Liaise with R&D to insure that product meets requirements and is

delivered in a timely fashion.

Coordinate with various internal marketing groups to insure that

appropriate marketing programs and collateral is produced.

Insure field readiness through development of field training programs

and delivery support and partner development.

By adopting a streamlined R&D and go-to-market processes, was able to

deliver three versions of a new product in 18 months and drive over

$10M in new revenue.

Sr. Systems Engineer September 1997 – December 2002

Utilized Solution Selling methodology to provide technical sales support

for both pre-sale and post-sale activities, helping to insure that regional

revenue objectives were obtained.

Activities included customer pain identification, formulating creative

technology solutions to address those pains, and then articulating the

value of the solution to various stakeholders in the organization through

customized proposals, presentations and demonstrations.



Also acted as mentor to new hires, as well as provide national leadership

on selected technology areas.

Subject Matter Expert—(J.G.)

Relevant Experience

SAS Institute, Inc. (1996 – Present)

Statistician July 2008 – Present

Designed and implemented advanced fraud discovery systems in the areas

of welfare, finance, and insurance.

Most recent project involved detection of social services fraud for a large

urban county.

• Project combines general anomaly detection rules, predictive modeling

techniques and social network analysis.

• Project is expected to eliminate tens of millions of dollars per year in

fraudulent welfare payments.

In other duties at SAS, has personally authored most of the company’s data

mining training curriculum.

Statistical Services Specialist 2002 – July 2008

Created and taught course in exploratory analysis, inferential statistics and

predictive modeling. Developed and maintained most of SAS Institute’s data

mining curriculum.

Analytic Consultant 1996 – 2002

Provided consulting, educational, and project management services to

clients. Specialization in predictive modeling, time series analysis,

exploratory analysis, data visualization and unsupervised classification

methods.

Statistics Unlimited, Inc. 1993 – 1995

Statistical Analyst

Engaged in the design and analysis of experiments in the biomedical,

environmental, defense, and chemical industries.

Information Arts, International 1991 – 1993

Software Developer

Created educational software in C for studying fractals, chaos and dynamical

systems.

Freelance Programmer 1982 – 1987

Developed and maintained software in medical equipment, manufacturing, financial

and educational domains.



Subject Matter Expert—(S.B.)

Relevant Experience

SAS Institute

Principal—Fraud Strategy (March 2009 – Present)

Serve as catalyst for providing SAS Fraud Framework solution direction to

maintain SAS’s leadership position within fraud and financial crimes.

Manage SAS fraud capability across industries at the global level.

Engage with key clients, prospects, and partners to ensure SAS Fraud

Framework keeps pace with emerging trends.

Serve as subject matter expert to support fraud organization design and

solution implementations.

Manage end-to-end delivery of enterprise SAS Fraud Framework.

Detica Consulting LLC

Senior Manager—Risk & Regulatory Business Unit (2006 – 2009)

Built Detica Chicago office from inception to prosperous 11 person team

focused on Financial Services clients. Initiated expansion of Detica US

practice into the Retail Financial Services sector through leveraging risk and

regulatory expertise and the Fraud Detection and Risk Management

platform. Led series of mission critical consulting engagements to address

clients’ most visible risk management problems.

Fraud Detection and Risk Management Platform – US Business

Development and Delivery

• Managed US business development activities for Fraud Detection

Platform in the banking and insurance industries – built annual pipeline

to $15M+.

• Managed pilot and implementation projects for Fraud Detection Platform

across industries, resulting in up to 10X return on investment year 1.

• Developed US centric fraud and social network analytics implementation

methodology for execution of software pilots and implementations and

the respective sales approach.

Client On-boarding and Management Target Operating Model for Global

Investment Bank

• Designed target operating model for an industry leading operations and

technology platform to manage risk of Client On-boarding and Account

Maintenance (COBAM) processes across global footprint.

• Established strategic roadmap for integration of workflow, case

management, regulatory rules, and document repository technology

components across all lines of business and support functions.

• Leveraged project expertise to develop Deice COBAM offering.

Regulatory Risk Platform Implementation for Global Banking Organization



• Managed software development lifecycle for custom development and

implementation of regulatory risk platform to support Client Lifecycle

Management, Anti-Money Laundering (AML) and ‘Know Your Customer’

(KYC) operations in North America.

• Facilitated Joint Application Development sessions across bank’s global

footprint, ensuring regional user requirements were appropriately

integrated.

PricewaterhouseCoopers LLP

Advisory Manager—Financial Services Risk and Regulatory Practice (2005 – 2006)

Developed PwC FS Risk and Regulatory practice into a market leader in the

Chicago Financial Services Industry. Established strong working

relationships with key banking and exchange/securities clients and became

their trusted advisor for consulting services. Facilitated sale of $10M in

consulting services and managed delivery of associated projects.

AML/KYC Implementation for Global Banking Organization

• Managed design of target operating model for client on-boarding process

in support of the Anti-Money Laundering (AML) and ‘Know Your

Customer’ (KYC) initiatives for the Client Engagement Group.

• Implemented 25 person operations group to support client on-boarding

processes. Defined roles and responsibilities and outlined a

management framework to drive operational efficiency.

• Identified key operational metrics and negotiated stakeholder

agreements with the businesses supported by the Client Engagement

Group.

Operational Risk Assessment for Global Banking Organization

• Facilitated series of Risk Assessment workshops for the Operational

Risk Management group of the bank’s Consumer and Commercial

Division focusing on financial reporting, KYC and AML operations, and

financial and operational effectiveness.

• Devised action plans for mitigating the operational risks identified in the

workshops.

Regulatory Reporting Remediation for Consumer Finance Unit of Global

Bank

• Managed cross functional team through the review of the regulatory

reporting process and designed a streamlined method for the analysis

and refiling of quarterly regulatory reports for the consumer finance unit.

• Identified opportunities to improve regulatory reporting processes and

formulated strategic plans for the implementation of the defined

enhancements.

• Designed and delivered a custom regulatory reporting training program

for client management and staff.



PGS Trading LLC

Owner and Risk Manager (2000 – 2005)

Expanded individual trading operations and founded PGS Trading LLC.

Developed business from inception to a multi-strategy investment fund

specializing in OTC and listed securities, futures, and options.

Built investment fund to eight employees and an annual revenue base of

$1,000,000.

Developed financial models and cash flow projections to determine the most

profitable trading methods to utilize.

Utilized quantitative and qualitative analysis to develop trading databases

and implemented computerized models to manage positions. Models

applied technical analysis, long-short pair trading, and arbitrage strategies.

Implemented risk management procedures to limit downside exposure of

positions.

Designed and conducted training and continuing education programs to

improve staff trading efficiency. Enhanced trader skills in arbitrage

identification and trade entry/exit optimization. Increased revenues by 25%

from inception of programs.

Developed business relationships and negotiated contracts with several

securities firms to cut trading costs by 40%.

Deloitte Consulting

Consultant—Financial Services and Healthcare Divisions (1997 – 2000)

Led teams of Deloitte and client personnel to further business development

and to enhance information technology systems. Utilized financial modeling

tools for system valuation and built consensus around strategic

recommendations.

Vendor Selection for Futures Clearinghouse

• Facilitated meetings with executives to define the business and technical

requirements of an enhanced clearing system.

• Executed vendor selection process using customized quantitative

methodology and conducted gap analysis to determine software and

hardware deficiencies between defined requirements and vendor

systems.

• Generated financial models and presented strategic recommendations to

executive board.

Design of Enhanced Clearing System for Options Clearinghouse

• Devised and implemented strategies for the design and development of

data distribution functions.

• Led team to create a new data structure and architecture platform for a

superior system.

Implementation of Financial System for Healthcare Provider



• Deployed testing approach and facilitated cross-division issue resolution

between accounting and IT leadership.

• Managed development and integration testing of electronic reporting

system. Improved automated acceptance of electronic reports by 55%.

Expertise

Seven plus years of management consulting experience in the across

industries with a focus on managing and delivering projects in fraud and

regulatory risk management platforms and operational strategy.

Built US pipeline for Fraud Detection and Risk Management platform from

scratch to $15M+ annually in insurance and banking industries within 1 year

timeframe.

Established reputation as expert in fraud analytics space, with regular

speaking engagements at industry conferences and executive roundtables.

Expanded consulting business into retail financial services (banking and

insurance) sectors through leveraging Fraud Risk platform.

Advanced Risk and Regulatory practice into a market leader in the Chicago

Financial Services Industry. Facilitated sale and delivery of $10M in advisory

services.

Demonstrated entrepreneurial and management capabilities by starting a

new business to capitalize on strategic opportunities within the securities

markets. Built revenues to $1M+ annually.

Project Manager—(C.B., PMP)

Relevant Experience

SAS June 2009 – Present

Project Manager

Manage client facing implementations of hosted SAS software solutions and

analytical services.

Experience implementing advanced analytical software and services for the

insurance, telecom and pharmaceutical industries.

Proven track record of high quality work delivered on time and on budget.

Microsoft Corp. (Microsoft Advertising) June 2007 – May 2009

Senior Project Manager

Most recently managed a 4200 man-hour project to successfully implement a

new Online Ad Server and two Yield Management tools for

Careerbuilder.com.

Utilized Agile Development principles to reduce project timelines and

increase customer involvement in final product.

Managed project financials and resource burn down with weekly reports to

management.



Successfully managed 35 Online Ad Server implementations, 6 licensed

upgrades, 60 hosted upgrades and 12 custom development project in my

first 6 months.

Integrated Salesforce into our revenue tracking and daily tactical operations.

Led the effort to fully map our Custom Development and Implementation

processes.

Mentored newly hired Project Managers.

Peopleclick, Inc. (April 2001 – June 2007)

Consulting Project Manager Jan 2007 – June 2007

Managed the implementations of a hosted Applicant Tracking System

(Peopleclick RMS) under the SaaS model.

Consulted on business practices and gathered system requirements

Acted as subject matter expert for other Project Managers for general system

knowledge specifically, Data Integrations with 3rd party vendors and Internet

Job Boards (Monster, Careerbuilder, etc.)

Business Analyst (informal Project Manager) Jan 2006 – Jan 2007

Mapped product enhancements against external client needs and internal

business requirements

Defined software requirements and system Use Cases for Peopleclick RMS

portal product including those enhancements designed for the emerging

High Volume market

Gathered requirements via customer interviews, JAD and RAD sessions with

Subject Matter Experts and cognitive walkthroughs.

Managed projects for the creation of a new internal environments and data

exchange scheduling modifications.

Implementation Consultant June 2003 – Jan 2006

Maintained a long term relationship with a subset of Peopleclick’s tier 1

accounts.

Lead an effort to more efficiently integrate the Peopleclick’s job posting

engine with common Job Boards (Monster, CareerBuilder, etc.) using SOA

principles.

Consulted with clients to ensure their requested changes met their business

need and to see if there was a more cost efficient method of achieving their

goals.

Gathered and documented requirements for Development, Quality

Assurance and Client Approval.

Member of a process design team responsible for defining how client

initiated Change Requests are completed accurately and in a timely manner.

Responsible for creation and maintenance of Change Request

Requirements Design Guide.



Quality Assurance Analyst April 2001 – June 2003

Performed user level validation for custom development prior to release.

Improved internal processes and overall client satisfaction.

Fraud Analytics Manager—(J.J.)

Relevant Experience

SAS

Director, Advanced Analytics Laboratory

Handles the customer business problems which require high-end analytics.

Current focuses in AAL are on credit scoring and fraud detections in

financial, health care, service and many other industries.

Samsung Life Insurance, Seoul, Korea

Vice President of CRM

Led three primary departments; CRM Infrastructure, CRM Strategy, and

CRM Operations to support the Sales and Marketing, Services, and Risk

departments.

• The areas involved: CRM Business Strategy, Data Collection Strategy,

and CRM Systems such as SFA, CMS, EDW and Call Center, Out-

Bound Call Center Execution, Market Segmentation, Customer’s lifetime

value, VIP program, Market & Consultant Research, and

Acquisition/Cross/Up-Sell/Retention/Risk Analytic & Business Model etc.

Experienced Analytical consulting with many of fortune 100 companies

including financial, telecommunication, retail, and pharmaceutical, etc.

• Type of Projects are Cross-sell/Up-sell, Fraud detection, Promotional

response model, Churn (Attrition), Fair lending detection, Drug

degradation (stability), Mail direct marketing response, Portfolio

optimization, Customer loyalty and life-time value, Default modeling,

Credit scoring, Domino effect, Campaign evaluation, Customer

segmentation, Scheduling, randomizing, and designing of experiments,

Customer satisfaction survey, Web-mining

Relevant Awards/Recognition

ASA Best Student Paper Award in Biopharmaceutical Statistic Section

(1997)

2001 SAS.COM magazine cover story (May/June) about the work with Bank

of America

2001 REALware award in e-CRM area with Bank of America



Fraud Analyst Lead—(S.A.)

Relevant Experience

SAS

Analytical Consultant

Fraud Detection in Information Technology

• Lead developer and implementer of a fraudulent access detection

system for a leading on-line information services provider.

• Helped design and develop the system that used weblog and other data

for monitoring usage patterns and other related activity. Many high end

analytical methods were used for alerting suspicious usage with

successful results.

Fraud Detection in Social Services

• Lead analytical consultant in a project related to detecting fraud and

abuse in social welfare.

• Helped develop and implement methods for detection and prevention of

fraud and abuse including business rules engine, anomaly detection

system, fraud prediction system and social network analytics to great

success.

Credit Scoring in Financial Services

• Served as analytical and business consultant in a number of SAS®/Credit Scoring engagements.

• Built acquisition and behavior scorecards for credit card, auto lending,

and mortgage portfolios in both production and pre-production

capacities.

Customer Retention Modeling in Banking

• Developed customer retention solutions for multiple banking

organizations.

• Developed models for ranking customers based on their propensity for

attrition. These analyses involved working with large datasets with large

number of variables, statistical segmentation, all aspects of predictive

modeling and supporting automated lead distribution as the end state of

the solution.

Fraud Application Developer—(J.K.)

Relevant Experience

SAS (2000 – Present)

Senior Analytical Consultant 2008 – Present

Achieved expertise in fraud detection, statistical process controls, customer

retention, cross/upsell modeling, customer acquisition, insurance rate

making, and credit modeling.



Analytical Consultant 2000 – 2008

General Dynamics 1998 – 2000

Metrics Engineer

ManTech Environmental 1997 – 1998

Statistical Programmer

DynTel 1996 – 1997

Programmer Analyst

Holcombe and Associates 1994 – 1996

Actuarial Assistant

Relevant Training

2007, SAS®9 Base Certified, SAS Institute

Fraud Analyst/Modeler

Relevant Experience

SAS Institute Inc.

Analytic Engineer (2009-present)

Insurance and health care fraud analyst/modeler. Implements fraud

detection rules; identifying anomalies that can help detect fraud, using of

advanced analytic and statistical techniques and using the social network

analysis to prevent fraud. Expert user of SAS Social Network Analysis and

has hands on experience in developing user guides and delivering user

trainings.

Accenture, Bangalore, India (2006-2008)

Operations Manager

Worked as an expert team member of Global Sales Support and Solutions

team for Business Process Outsourcing & Project Mobilization, with key

expertise in solution, proposal response development and bid support.

Tata Interactive Systems, Mumbai, India (2005-2006)

Manager-Corporate Marketing Group

Managed a cross functional team of consultants focusing on planning of

large to medium sized e-Learning solutions. There he developed project

plans including effort estimations and costing.

IBM Global Services, Gurgaon, India (2001-2005)

Sr. Executive-Business Development and Customer Operations



Education

Masters of Science in Analytics (2009)

North Carolina State University, Raleigh, NC

MBA-Information Systems & Marketing (2001)

GGS Indraprastha University, Delhi, India.

BA-Economics, Education, English and Modern Indian Language (1999)

University of Delhi, Delhi, India

Relevant Skills and Training

SAS Social Network Analysis

Certified Advanced Programmer for SAS®9

Certified Base Programmer for SAS®9

Predictive Modeler using SAS® Enterprise Miner™ 5

Six-Sigma Black Belt

ETL Development Lead—(B.M.)

Relevant Experience

SAS (June 1986 – Present)

Software Consultant May 2004 – Present

Designed and built a campaign performance report system. Used Enterprise

Guide and Data Integration Studio to build an ETL process combining SAS

campaign reference data, customer contact data and campaign

response/revenue data. Resulting data is consumed by a SAS Information

Map used in SAS Web-based reports. The report displays key performance

indicators related to the effectiveness of the campaign.

Directed and lead a team of SAS consultants and customer resources to

deliver C-level executive reports to help uncover customer sales promotion

effectiveness. Designed and built SAS On-Line Analytic Processing (OLAP)

cubes using SAS OLAP Cube Studio. Produced and deployed web-based

reports for the SAS Information Delivery Portal using SAS Information Map

Studio and SAS Web Report Studio.

Selected from a global SAS consulting organization to attend SAS

Telecommunications Intelligence Solutions training sessions in Heidelberg,

Germany and Miami, Florida to facilitate future sales opportunities related to

telecommunications analytics. Each session covered installation and usage

of the SAS Intelligence Solution package. The package includes pre-defined

Detail Data Stores, Analytic Base Tables and Analytic Data Models.

Team co-lead for a customer-facing SAS Telco Intelligent Solution Technical

Assessment. Reviewed existing database tables to perform a gap-analysis

for the matrix-of-fields detail data store document.



Performed SAS 9.1.3 SAS Software installations and knowledge transfers at

customer sites. Produced customer deliverable installation documents.

Installations components include Xythos WFS, Apache WFS, Jakarta

Tomcat 4.1.18, BEA Weblogic, SAS mid-tier components, SAS Forecast

Studio, SAS EMiner 5.2, SAS DI Studio, SAS Enterprise Guide 4.0. Tested

the start-up functionality of each SAS software application. Software

installations performed on UNIX and Windows Operating Systems.

Assessed Windows Server Sizing and Usage Assessment for SAS BI Server

architecture. Microsoft-based PERFMON utility used to capture performance

metrics over a 24 hour time interval. Utilized Process Monitor from

sysinternals.com for similar server performance engagements.

Presented impromptu white-board presentations of the SAS 9.1.3

architecture design to educate customers during pre-sales engagements.

Demonstrated SAS 9.1.3 ETL/DI Studio product overview presentations for

pre-sales and post-sales engagements.

Software Manager/Systems Developer July 1999 – May 2004

Project Manager for SAS/Warehouse Administrator Add-ins Development

Team

Managed six programmers whose collective tasks involved the development,

testing, and release engineering for add-ins developed for releases 2.0, 2.1,

and 2.2

Responsible for the design, implementation and customer release of each

add-in release

Extensive use of Microsoft FrontPage web page development tools

Monitored and used internal SAS Institute OASIS project management tools

Developed Target Designers and AboutBox versioning information for

Version 9 SAS DataBuilder Java Client software.

Responsible for EDM divisional roll-out of the pilot project usage of Oracle

ITime

Responsible for initial investigation and implementation of Microsoft Project

Software used for EDM project management.

Systems Developer June 1998 – July 1999

Helped develop SAS Institute Defects Data-Warehouse. Scheduled

warehouse load jobs.

Maintained servers that hosted the warehouse databases.

Maintained and extended functionality of existing SAS Warehouse

Administrator Release 1.3 add-ins.

Project Manager June 1994 – June 1998

Project Manager of Release Engineering and Source Management Team

within Open Systems R&D Host Division.



Managed five programmers whose collective tasks involve the coordination

of the release and source management of the 6.09, 6.10, 6.11, and 6.12

SAS systems for UNIX based machines.

Responsible for the design and implementation of the Version 7 SAS

installation software for UNIX platforms.

Provided direction to the Version 7 SAS system divisional source

management research groups.

Systems Developer June 1991 – June 1994

Developed configuration management tools for the Open Systems R&D Host

Division.

Member of the porting process team that designed and implemented

software used to port the SAS system source code to UNIX platforms.

Responsible for building and debugging the SAS system across five different

UNIX platforms.

Member of the UNIX SAS installation design team. Coded shell script

modules used in the installation software.

Worked with the manufacturing department to prepare the 6.09 SAS system

for shipment to customers.

SAS Associate Systems Developer June 1986 – June 1991

Developed the source management and configuration management tools

used to port six million lines of the SAS system source code for the Data

General/Prime Host Group.

Responsible for compiling and linking the SAS system.

Worked with other host groups within the Host R&D division in setting

standards for porting and building the SAS system.

Installed the SCO UNIX operating system on Intel 486 PCs; and ported and

built the SAS system for this platform.

Analytical Consultant—(E.B.)

Relevant Experience

SAS Institute, Inc.

With over 25 years of experience, E.B. has repeatedly demonstrated strengths in

statistical data analyses and analytical consulting, with a successful career in

coaching customers to run data mining projects.

For 11 years, doing business oriented consulting specialized in SAS data

mining technologies. Experience in supporting customers to plan and

develop predictive models with statistical rigor.

Mentors and supports SAS products and solutions used in the industry.



Demonstrates strong skills set as instructor, writer and presenter in the

medical and academic fields.

Recent Project Experience

Provided epidemiological expertise for data analyses and mapping with the

Federal Accounts associated with cancer and health.

Has coached hospital customers in use of Enterprise Miner and SAS/STAT,

helped model assessment and fine tuning when estimating disease rates

Scoped, designed and crated a logical model to estimate fraudulent claims in

an unemployment data bases;

Done many pre-sales projects in health care area.

Developed projects in SAS forecast server and Enterprise Miner to assist the

prediction of enrollment drop-off rates in a health insurance agency

Scoped and developed a SOW to estimate claims likely to appeal to the

federal district court.

Previous employers include:

Duke Clinical Research Institute, where she worked as a Senior Statistician

for 2 years, in charge of statistical consulting involving research design and

sample size calculation. She performed in-depth statistical analyses in large

confidential data sets from clinical trials.

Comprehensive Cancer Center , where she developed analytical models in

SAS and performed statistical analyses in a population-based case-control

study.

Maternal Child Health Department, UNC Chapel Hill where she worked as

statistician, analyzing large National Center for Health Statistics databases to

calculate rates of birth outcomes and prenatal care utilization for the

southern US.

San Diego State University Foundation project HELPS. She worked as

research associate and supervised personnel, developed data management

systems and BMDP/SAS programs to manipulate data on longitudinal

hypertension study.

Degrees and Credentials

Obtained PhD in Epidemiology from UNC Chapel Hill, Chapel Hill, NC

Obtained MPH in Biostatistics from San Diego State University, San Diego,

CA

Obtained BS in Statistics from the State University of Campinas-Brazil



SAS Engagement #1: Washington State Department of Labor and Industries

F. For responders submitting proposals to subject matter categories #1

through #11, a detailed description of data analytics services provided to two

or more public sector organizations of similar size and scope to the State of

Minnesota within the past three years (“engagements”), specific to the

category(ies);

In Washington State, the Department of Labor and Industries (L&I) administers the

state-operated workers’ compensation. L&I is the eighth largest workers’

compensation insurance company in the country providing coverage for more than

2.5 million workers employed by 171,000 employers. This “State Fund” provides

workers with wage-replacement and medical benefits to offset the financial impact of

a job-related injury or occupational disease. This no-fault insurance protects

employers from lawsuits when work-related injuries and diseases occur. Premiums

paid by employers and workers finance the State Fund. L&I is one of four exclusive

State Funds in the nation. Each year, the State Fund collects in excess of $1.4

billion in premiums and receives approximately 137,000 new claims.

WA L&I purchased the SAS Fraud Framework for Government to address workers’

compensation premium evasion by detecting unregistered employers that are not

paying workers’ compensation taxes for their employees. In addition, it will uncover

the most serious cases of underreporting by employers that are part of its system.

This will enable the agency to best focus its audit resources, increasing the return on

investment (ROI) from every staff person aimed at preventing premium evasion.

WA L&I is in the middle of implementing the solution, with two phases in place. The

first phase provides advanced identity resolution as well as a comprehensive

employer profile and analytic screens for staff who review potential detection hits.

The second phase built predictive models based on data from more than a dozen

different systems across five separate Washington State agencies. The final two

phases are nearing completion, and will be in place by the end of June 2011. That

final phase includes IRS data in the predictive models, as well as integration with our

auditing system.

The SAS Fraud Framework for Government is the cornerstone of the agency’s fraud

detection and prevention solution, leapfrogging the detection systems built in house.

WA L&I expects the SAS solution to fully pay for itself within 12-18 months of

implementation using conservative estimates, and improve on its current ROI of $8

for every $1 spent. For implementation, WA L&I focused on workers’ compensation

premium evasion, but expect to find valuable hits for its state unemployment

insurance and revenue agencies as well, as their data is included within the system.

WA L&I plans to share those hits, further increasing the potential return.

8th largest commercial

insurance company in

the US

Maximizing existing

resources to further

increase ROI

Integration of data from

a myriad of sources,

including IRS data

Solution will pay for

itself



G. For responders submitting proposals to subject matter categories #12, a

detailed description of data analytics services provided to two or more public

or private sector organizations of similar size and scope to the State of

Minnesota within the past three years (“engagements”), specific to the work

intended under this category;

Not Applicable.

H. For each engagement, provide contact information for the individual

responsible for the project from each public sector organization (or private

sector for category #12, if applicable) that can provide additional information

regarding the project and verify any representations made. Specifically,

include the name of the organization; name and scope of project; dates of

engagement; and name of contact person, including the individual’s e-mail

address and direct telephone number.

Washington State Labor and Industries (L&I)

Workers’ Compensation Premium Evasion Project

December 2009 - Present

Carl Hammersburg

Fraud Prevention and Compliance

Manager

[email protected]

(360) 902-5933

I. For each engagement, describe the efficiencies or results gained by

implementation of the data analytics and, if applicable, any cost savings

determined to be directly attributable to the project.

The agency expects the SAS solution to fully pay for itself within two years of

implementation using conservative estimates. For implementation, the agency is

focusing on workers’ compensation premium evasion, but expects to find valuable

hits for the state unemployment insurance and revenue agencies as well, as the data

that is included within the system. WA L&I plans to share those hits, further

increasing the potential return.

Currently, the department is uncovering about $26 million of the premium revenue

lost annually and is realizing a return on investment of about eight to one. In other

words, for every dollar WA L&I spends investigating fraud, it gets eight back. Along

with revenue recovery, the solution will make investigative work more efficient. As an

example, a basic information search on an employer will be reduced from 30 minutes

to 30 seconds. In 2009 alone, WA LNI collected more than $137 million from such

things as claim overpayments and fraud and delinquent employer premiums and

fraud, and saved the agency another $10 million in what would have been ongoing

costs related to fraud. It has already resulted in a dramatic drop in the time it takes

to research a case—within 30 days it went down by 80 percent.

Leveraging information

for other state agencies

8-to-1 return on

investment and 80%

increase in efficiency

mailto:[email protected]



J. For each engagement, describe how any such cost savings were calculated

and how compensation for the work was structured.

Cost savings were calculated by the customer. SAS received compensation through

a standard software licensing contract and a professional services contract with

payment terms of net 30 days.

K. For each engagement, detail whether any disagreements as to savings

generated or fees owed were encountered and if so, how resolved.

Not applicable.

L. For each engagement, to the extent compensation was based on a

percentage of savings attributable to the effort; describe how the parties

distinguished the amount of savings attributable to the work of the contractor

versus savings realized from the efforts of the organization.

Not applicable.

SAS Engagement #2: Louisiana Workforce Commission

F. For responders submitting proposals to subject matter categories #1

through #11, a detailed description of data analytics services provided to two

or more public sector organizations of similar size and scope to the State of

Minnesota within the past three years (“engagements”), specific to the

category(ies);

Historically, Louisiana has been a hotbed of fraud in the U.S. A commonly retold

joke is that half of the state is underwater and the other half is under indictment. The

state’s approach to detecting and mitigating fraud was fragmented. Agencies and

programs did not communicate or share information–or at all, in many cases. The

Governor wanted Louisiana to be a desirable place to move business and/or expand

business, but the effect that fraud has on businesses makes the state a less-than-

appealing destination. The state lacked the necessary resources and technology

infrastructure to identify, manage and mitigate fraud. By not addressing this issue,

the state would continue to lose business and opportunities to other states and would

continue to have to borrow money from the federal government to prop up depleted

trust funds.

The state is taking an enterprise approach to fraud management using SAS Fraud

Framework for Government. SAS heavily emphasized fraud’s relationship to the

efficiency and solvency of government. Addressing fraud on a statewide level could

directly impact tax rates, bond ratings, and economic development, and help reduce

budget deficits. Louisiana is currently involved in a phase one project to address

fraud in its unemployment insurance (UI) and workers’ compensation insurance

programs. The long-term vision is to aggregate the state’s data and expand its fraud

management efforts into other areas, including temporary assistance for needy

Louisiana chose SAS as

its comprehensive

strategy to address

fraud across all state

programs

WC and UI are the first

programs addressed



families (TANF), supplemental nutrition (SNAP), child support enforcement, child

care, Medicaid, Medicare, tax and revenue.

The Advanced Analytics Lab employs scientists and domain experts to create fraud

models for Louisiana. SAS Solutions OnDemand hosting solution provided a way to

mitigate risk and gain faster time to value. In age of economic and budget crises,

Louisiana did not have to purchase more hardware or hire more people to manage

fraud – it just partnered with SAS. As part of the contract and deliverables, SAS is

executing a quick-start program with the customer by performing manual analysis

against the state’s data to deliver results in less than 100 days. At same time, SAS is

building a more permanent infrastructure for long-term fraud analysis. The quick start

and robust infrastructure were attractive aspects of the proposal..

The SAS solution is breaking down barriers and silos of data and facilitates the

sharing of information across agencies and boundaries. As a result of the quick start

program, Louisiana received a check via FedEx for $454,000 from an offending

company without a fight. In the long term, the state estimates an 8-to-1 return on its

investment in SAS. Based upon documented success in other projects, SAS feel this

ratio is very conservative and could go much higher. Efficiency gains are also

realized in the automation of investigative activities. What has historically taken

approximately 86 hours has now been reduced to less than 6 hours.

Louisiana’s efforts around fraud will also produce the first real instance of a citizen-

state vector by essentially create a master person index through identity resolution

and other Master Data Management solutions. If “Person A” is a Louisiana citizen,

the government could pull up a unique identification code and look at the program

effectiveness, risk score, productivity index and other metrics of that individual

citizen. Taking a longitudinal view of a student in terms of their educational life cycle

is very common. It’s not common in other areas of government, but there are many

interesting benefits for the state and the citizens that can be derived from it.

G. For responders submitting proposals to subject matter categories #12, a

detailed description of data analytics services provided to two or more public

or private sector organizations of similar size and scope to the State of

Minnesota within the past three years (“engagements”), specific to the work

intended under this category;

Not Applicable.

H. For each engagement, provide contact information for the individual

responsible for the project from each public sector organization (or private

sector for category #12, if applicable) that can provide additional information

regarding the project and verify any representations made. Specifically,

include the name of the organization; name and scope of project; dates of

First results in less than

100 days

$454,000 check

received without any

litigation required

Citizen-state vector or

master person index in

the future



engagement; and name of contact person, including the individual’s e-mail

address and direct telephone number.

Louisiana Workforce Commission

First fraud project addressed by the Statewide Fraud Detection System

August 31, 2010 - present

Wes Hataway, Director, Workers Compensation

[email protected]

225-252-0578

I. For each engagement, describe the efficiencies or results gained by

implementation of the data analytics and, if applicable, any cost savings

determined to be directly attributable to the project.

As a result of the quick start program so far, Louisiana received a check via FedEx

for $450,000 from an offending company without a fight. In the long term, the state

estimates an 8-to-1 return on its investment in SAS. Furthermore, time dedicated to

investigations was reduced from 85 hours to less than 6 hours, approximately.

J. For each engagement, describe how any such cost savings were calculated

and how compensation for the work was structured.

Cost savings were calculated by the customer. SAS received compensation through

a standard software licensing contract and a professional services contract with

payment terms of net 30 days.

K. For each engagement, detail whether any disagreements as to savings

generated or fees owed were encountered and if so, how resolved.

Not Applicable.

L. For each engagement, to the extent compensation was based on a

percentage of savings attributable to the effort; describe how the parties

distinguished the amount of savings attributable to the work of the contractor

versus savings realized from the efforts of the organization.

Not Applicable.

State expects and 8-to-

1 ROI; investigative time

reduced from 85 to 6

hours

mailto:[email protected]

4.6 Workers’ Compensation - WordPress.com · SAS has been partnering with customers to solve...

Documents

Transcript of 4.6 Workers’ Compensation - WordPress.com · SAS has been partnering with customers to solve...